Characterisation and optimisation of the semiconductor optical amplifier for ultra-high speed performance

Shalaby, A. A. (2012) Characterisation and optimisation of the semiconductor optical amplifier for ultra-high speed performance. Doctoral thesis, Northumbria University.

PDF (PhD thesis)
shalaby.ahmed_phd.pdf - Accepted Version

Download (19MB) | Preview


This research is in the area of high speed telecommunication systems where all- optical technologies are being introduced to meet the ever increasing demand for bandwidth by replacing the costly electro-optical conversion modules. In such systems, all-optical routers are the key technologies capable of supporting networks with high capacity/bandwidth as well as offering lower power consumption. One of the fundamental building blocks in all-optical routers/networks is the semiconductor optical amplifier (SOA), which is used in for clock extraction, wavelength conversion, all-optical gates and optical processing.
The SOAs are perfect for optical amplification and optical switching at a very high speed. This is due to their small size, a low switching energy, non-linear characteristics and the seamless integration with other optical devices. Therefore, characterisation of the SOA operational functionalities and optimisation of its performance for amplification and switching are essential and challenging. Existing models on SOA gain dynamics do not address the impact of optical propagating wavelength, the combined input parameters and their adaptation for optimised amplification and switching operations. The SOA operation is limited at high data rates > 2.5 Gb/s to a greater extent by the gain recovery time. A number of schemes have been proposed to overcome this limitation; however no work has been reported on the SOA for improving the gain uniformity.
This research aims to characterise the boundaries conditions and optimise the SOA performance for amplification and switching. The research also proposes alternative techniques to maximise the SOA gain uniformity at ultra-high speed data rates theoretically and practically. An SOA model is been developed and used throughout the research for theoretical simulations. Results show that the optimum conditions required to achieve the maximum output gain for best amplification performance depends on the SOA peak gain wavelength. It is also shown that the optimum phase shift of 180º for switching can be induced at lower input power level when the SOA biasing current is at its maximum limit. A gain standard deviation equation is introduced to measure the SOA gain uniformity. New wavelength diversity technique is proposed to achieve an average improvement of 7.82 dB in the SOA gain standard deviation at rates from 10 to 160 Gb/s. Other novel techniques that improved the gain uniformity employing triangular and sawtooth bias currents, as replacements for the uniform biasing, have been proposed. However, these current patterns were not able to improve the SOA gain uniformity at data rates beyond 40 Gb/s. For that reason, an optimised biasing for SOA (OBS) pattern is introduced to maximise the gain uniformity at any input data rates. This OBS pattern was practically generated and compared to the uniform biased SOA at different data rates and with different input bit sequences. All executed experiments showed better output uniformities employing the proposed OBS pattern with an average improvement of 19%.

Item Type: Thesis (Doctoral)
Uncontrolled Keywords: SOA gain uniformity improvement, output gain standard deviation, wavelength diversity technique, triangular and sawtooth biasing currents, optimised biasing for SOA
Subjects: H600 Electronic and Electrical Engineering
Department: Faculties > Engineering and Environment > Mathematics, Physics and Electrical Engineering
University Services > Graduate School > Doctor of Philosophy
Related URLs:
Depositing User: Ay Okpokam
Date Deposited: 21 Mar 2012 16:12
Last Modified: 17 Dec 2023 14:06

Actions (login required)

View Item View Item


Downloads per month over past year

View more statistics